J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.32.6.530 on 1 December 1969. Downloaded from

J. Neurol. Neurosurg. Psychiat., 1969, 32, 530-540

Saphenous conduction in man1

CUMHUR ERTEKIN2 From the Laboratory ofNeurophysiology, University Hospital, Lund, Sweden

At present, the most reliable methods for inves- four females) were 41 to 63 years (mean age 51). In tigating sensory nerve conduction for diagnostic control subjects, 53 saphenous were studied and, purposes involve the median and the ulnar nerves in addition, eight subjects underwent measurement of the (Gilliatt and Sears, 1958; Buchthal and Rosenfalck, distal sensory conduction in the posterior . Studies of the distal sensory conduction of the median 1966). In the lower extremities, pure sensory con- nerve were also made. duction is, however, not easily studied. It has Studies of the sensory conduction in the saphenous recently been shown that sensory potentials recorded nerve were carried out in 65 patients, 36 of whom (31 in the posterior tibial and the fibular nerves at the males, five females) showed clinical signs of polyneuro- have a considerably lower amplitude than the pathy and/or associated slow motor or sensory conduc- sensory action potentials in the nerves of the arms tion in some other nerves. The mean age of these cases and that an averaging computer is necessary to was 51, ranging from 20 to 76 years. Twenty-nine of the obtain accurate recordings (Mavor and Atcheson, patients (15 males, 14 females) did not show any clinical guest. Protected by copyright. 1966; Buchthal and Rosenfalck, 1966). signs and had no subjective symptoms of polyneuro- pathy. They suffered from diabetes mellitus, chronic The difficulties mentioned in the study of the alcoholism, renal insufficiency, or other diseases which sensory conduction in the lower extremities are of may affect peripheral nerves. These 29 cases have for the great clinical disadvantage, since many polyneuro- sake of brevity been called the 'subclinical' group. The pathies first give rise to sensory symptoms in the legs. mean age of these cases was 36 (19 to 64) years. It is, therefore, often important to be able to measure Sensory action potentials were recorded both from the early changes of the sensory nerve in the legs, posterior tibial and the median nerves. The digital nerves especially in cases with only subjective sensory of the first and the first finger were stimulated and complaints and in which the EMG and the motor the action potentials were recorded at the ankle and the conduction in the is normal. wrist respectively by the method described by Buchthal legs and Rosenfalck (1966). In the present study, a technique is described to The saphenous nerve is a terminal sensory branch ofthe measure the sensory conduction in a nerve not and it innervates the medial surface of the studied previously in the lower extremity, the leg from the to the (Fig. 1). Above the knee, it saphenous nerve. This nerve proved to be easy to passes deeply within the subsartorial canal inthethighand locate and it always gave sensory action potentials arrives at the level of the where it joins in normal subjects on adequate stimulation. The other motor and sensory branches of the femoral nerve. conduction velocity values obtained from the In the canal as well as at the ligament, it lies close to the saphenous nerve will be compared with those found motor branches to the vastus medialis muscle, Just under in other nerves in normal subjects and in patients the inguinal ligament it is situated laterally to the point of the maximal pulsation of the . It is easy with polyneuropathy. to reach it by a needle electrode at this site. http://jnnp.bmj.com/ The following standard procedure was used for the MATERIALS AND METHODS studies of sensory conduction in the saphenous nerve. Stimulation of the sensory nerves was performed with Fifty normal subjects served as controls for the saphenous surface electrodes of the same type as those used for studies. The majority of these were healthy volunteers, stimulating the distal nerves (DISA 13K65). The two while others suffered from traumatic nerve injuries of the electrodes were mounted in parallel, perpendicular to the upper extremities only, but showed normal sensory and long axis of the leg just below the lower edge of the motor findings in the lower extremities. These controls on the medial surface of the were leg. They placed on October 2, 2021 by were divided into two groups according to age. at a distance of 2 to 3 cm and held in place firmly by Thirty-one subjects (23 males, eight females) were surgical tape. Before application of the electrodes the between 17 to 38 years (mean age 26) and 19 (15 males, skin was cleaned with ether and gently rubbed with 'Presented at the twenty-fourth annual meeting of the Swedish sandpaper. Medical Association in Stockholm on 1 December 1967. Electrical stimulation at this site has the advantage of 'On leave of absence from the Department of Clinical Neurology, not activating muscle tissue around the electrodes and Aegean University Medical School, Izmir, Turkey. hence artefacts from muscles are avoided. The proximal 530 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.32.6.530 on 1 December 1969. Downloaded from

Saphenous nerve conduction in man 531 FIG. 1. Diagram ofthe discarded. At the end of the study the motor fibres were saphenous nerve and the stimulated supramaximally-that is, with a stimulus 10L3 position ofthe electrodes intensity of five to 10 times the motor threshold-and the /A .. ~Le for stimulation and re- response in vastus medialis muscle was recorded. cording. The dotted line When the recording electrode at the inguinal ligament UC.- I S;_shows the course of the had been placed as close to the saphenous nerve as saphenous nerve and possible, the electrodes were connected to the input the solid line indicates transformer (ratio 1:30, 10 to 500 c.p.s.) of the the motor fibres of the electromyograph (DISA 14A30). The differential am- femoral nerve to the plifier of the electromyograph has an input impedance vastus medialis muscle. of 200 MD shunted with 15 pF, a lower frequency limit ST1 andST2:stimulation (3 dB down) at 2 or 20 c.p.s. (through high pass filter) electrodes for the and an upper frequency limit (3 dB down at 10,000 \ g saphenous nerve. REC: or 2,000 c.p.s. through a low pass filter) with a slope \ 1l site of the recording oft dB/octave. The noise level of the amplifier with short electrodes at the circuited input was less than 1-5 uV RMS, 7 ,uV peak to ST1 inguinal ligament. peak. The saphenous nerve was stimulated on the medial aspect of the knee and in some instances just above the medial malleolus with rectangular pulses 0-2 msec in duration and with an intensity of 40 to 60 mA (DISA Ministim 14 E 10). The stimulus current was monitored ST1 through a 1OQ resistance on one of the channels of the electromyograph. The sweep speed used for recording the sensory action guest. Protected by copyright. potential on the film was 0-5 msec/mm or, in some patients with polyneuropathy or in those stimulated at the medial malleolus, 1-0 msec/mm. Twenty superim- posed sweeps, as well as single sweeps, were photographed. The sensory conduction time was measured from the stimulating electrode was connected to the cathode of the beginning of the stimulus artefact to the peak of the first stimulator. positive deflection. The distance between the inguinal A concentric needle electrode (DISA 13K03) was ligament and the knee was then divided by the conduction placed in the vastus medialis muscle. time and the conduction velocity was expressed in in/sec. The electrodes used for recording sensory action In 10 normal subjects conduction time was obtained by potential were stainless steel needles, Teflon coated stimulating both at the knee and the medial malleolus. except at the tip. The stigmatic electrode had a bare tip The amplitude of the sensory potential was measured of 3 mm and the indifferent a bare tip of 5 mm. In order to from peak to peak. The motor nerve conduction time to reduce the impedance of the electrodes, an alternating the vastus medialis muscle was measured from the current was passed through each electrode for about 30 beginning of the stimulus artefact to the onset of the to 40 seconds, with the electrode placed in hot saline evoked muscle response, the value peak to peak of the (900). This procedure produced electrodes with an first high positive-negative deflection was considered as impedance of 1 to 2 KQ at 10 to 5,000 c.p.s, measured its amplitude. In every study the deep temperature at the with a peak to peak voltage of 50 pV. inguinal ligament was measured by an electrothermometer The recording electrodes were then connected to the (Ellab-Copenhagen) and found to be 360 to 37°C in both output of the stimulator (DISA Ministim 14 E 30). The the normal subjects and the patients. stigmatic electrode was inserted at the inguinal ligament http://jnnp.bmj.com/ about 0 5 to 1-5 cm lateral to the maximum pulsation of RESULTS the femoral artery, while the indifferent was inserted 1-5 to 3 0 cm more laterally. A long circular ground NORMAL SUBJECTS In Table I the results from the electrode was placed between the electrodes at the inguinal studies of sensory conduction in the saphenous ligament and in the vastus medialis muscle. With a stim- nerve and the motor conduction in the femoral ulus duration of 0-2 msec the position of the electrode nerve are summarized. Figure 2 shows examples of was then adjusted to a position which gave the lowest sensory action potentials in the saphenous nerve threshold for the motor response in the vastus medialis (left) and evoked motor responses in m. vastus med. muscle. The mean threshold for obtaining the muscle at high (middle) and low gain (right) in the low (A) on October 2, 2021 by action potential was found to be 0-6 mA. In a few patients the the threshold was about 1 0 mA or slightly higher. In all and high (B) age group. cases the motor threshold was also determined after Figure 3 demonstrates the sensory responses at the recording the sensory action potential. The mean thres- inguinal ligament in a 21-year-old normal man to hold was then found to be 0-7 mA. If the threshold at the supramaximal stimulation at different levels of the end of the study was higher than 1 0 mA the results were saphenous nerve. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.32.6.530 on 1 December 1969. Downloaded from

532 Cumhur Ertekin TABLE I SUmmARY OF RESULTS FROM NORMAL SUBJECTS Sensory conduction in saphenous nerve Age (yr) Mean Nerves Sens. cond. Sens. cond. Distance Amplitude age examined velocity time (msec) (cm) of sens. resp. (no.) (m/sec) ("sV) Young normal subjects (knee-ing. lig) 17-38 26 33 59 6 ± 2-3 7 0 ± 0 5 42-2 2-3 4-2 1-8 Old normal subjects (knee-ing. lig) 41-63 51 20 57-1 ± 2-3 7-3 ± 0 5 41-4 3-1 3-6 2-4 Young normal subjects (knee-ing. lig) 17-36 25 10 59 4 ± 2-8 7-1 + 0-3 42-0 1-2 4-8 2-4 (ankle-knee) 52-3 ± 2-3 26-6 ± 38 (ankle-ing. lig) 56-6 ± 2-0 68-8 3-8

Motor conduction in femoral nerve Age (yr) Mean Nerves Motor conduction time Distance Amplitude age examined to m. vastus med. (cm) of muscle (no.) (msec) resp. (m V) Young normal subjects 17-38 26 34 5 05 + 0 5 28-8 2-3 23-4 11-3 Old normal subjects 41-63 51 20 5-08 ± 0 5 27 5 3-1 21-5 ± 9.9

All values are mean ± S.D. guest. Protected by copyright.

Saphenous Femoral nerve nerve 60 M/sec 27 cm RC-. A 1

-- ] ]8pv ]2mV I4OmV B

54 M/sec 25cm _- 10 msec FIG. 2. Normal sensory action potentials from the saphenous nerve recorded at the inguinal ligament after

stimulation of the medial aspect of the knee in two healthy http://jnnp.bmj.com/ subjects. The middle and right traces show motor responses : evoked from the vastus medialis muscle by supramaximal ST, stimulation at the inguinal ligament both at high and low amplifications. All sensory responses are 20 superimposed potentials. The figures for sensory action potentials give the conduction velocity between the knee and the inguinal seI ligament and the figures for the motor responses show the 5 msec distances between the inguinal ligament and recording the sites in the vastus medialis muscle. A. Y.F., 27-year-old FIG. 3. Normal sensory action potentials from on October 2, 2021 by female (6 December 1967). The distance between knee and saphenous nerve recorded at the inguinal ligament (Rec) inguinal ligament is 42 cm. B. S.C., 52-year-old male after stimulation at the knee (ST1) and above the medial (11 September 1967). The distance between knee and malleolus (ST2) in a healthy subject. A.N., 21-year-old inguinal ligament is 405 cm. (In this figure and in all male (18 January 1968). The distance between ST1-ST2 is subsequent figures, a downward deflection of the trace is 252 cm and ST1-Rec 42-1 cm. All sensory responses are positive.) 20 superimposedpotentials. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.32.6.530 on 1 December 1969. Downloaded from

Saphenous nerve conduction in man 533

The mean sensory conduction velocity in the groups are given in Table II. Figure 4 gives some saphenous nerve between the knee and the inguinal examples from such comparisons. In the group ligament was 59-6 m/sec in young and 57-1 m/sec in where sensory conduction in the saphenous nerve old normal subjects. The mean amplitude of the was compared with that in the posterior tibial nerve, sensory action potentials recorded at the inguinal all eight subjects had detectable saphenous nerve ligament after stimulation at the knee was 4-2 ,uV sensory potentials but only six of them showed a in the young and 3-6 ,uV in the old group. The shape detectable sensory response at the ankle on stimu- of the sensory action potentials was in most instances lation of the digital nerves of the first toe. Moreover, triphasic but many of the cases had in addition smaller sharp late deflections. The difference in sensory conduction between the two groups was A B C significant (P<0-001). Below the age of 30, sensory 40.6 cm 39.9 cm 40.0cm conduction velocity in all normal subjects was above %, Saphenous ! ] d t] 55 m/sec but some subjects over 50 years had con- [ 01o_^,J nerve duction velocities below 55 m/sec, but none was recorded below 52 m/sec. The amplitude of the 14.0cm sensory action potentials and the distances between Median stimulation and recording sites did not differ ST. E, > ;. nerve significantly in the two different age groups. In 10 normal subjects (age 10 to 36 years), the 208cm 19.4cm sensory conduction velocity in the lower segment Posterior tibial

medial malleolus-knee of the saphenous nerve was guest. Protected by copyright. nerve 52-3 m/sec. This issignificantly(P < 0-001)slowerthan in the segment the conduction velocity proximal 5 msec. (knee-lig. ing.). The mean sensory conduction along the whole length of the leg was 56-6 m/sec. FIG. 4. Sensory action potentials recorded from the In eight normal subjects with a mean age of 45 saphenous nerve at the inguinal ligament (upper traces) years (33 to 59 years), sensory conduction was and the posterior tibial nerve at the ankle (lower trace) in investigated both in the proximal segment of the two healthy subjects (A, B) and from the median nerve at saphenous nerve and in the posterior tibial nerve the wrist (middle trace) in a healthy subject (C). All sensory responses are 20 superimposed potentials. Figures (first toe-ankle segment). In seven other normal on each trace give the distances between stimulation and subjects sensory conduction was determined both in recording sites. A. M.G., 36-year-oldfemale (7 July 1967). the proximal segment of the saphenous nerve and B. S.E.J., 37-year-old female (14 July 1967). C. K.H., in the median nerve (stimulation of the thumb 28-year-old female (18 January 1968). Vertical scales recording at the wrist). The results from these two indicate 8 m V. Horizontal scales indicate 5 msec. TABLE II COMPARISON OF SENSORY CONDUCTIONS IN NORMAL SUBJECIS (A) Between posterior tibial and saphenous nerve" Age Mean Sensory conduction Sensory conduction Distances Amplitude of No. recor- (yr) age veloc. first toe-ankl time first toe-ankle (cm) sens. responses dable sens.

or knee-inguinal or knee-inguinal at ankle or at potentials http://jnnp.bmj.com/ ligament (mlsec) ligament (msec) ing. lig. ("iV) (no. ofnerves) Posterior tibial nerve 33-59 45 35-7 + 3-8 5-8 06 205 1-1 1-33 04 2 Saphenous nerve 33-59 45 57-2 + 2-3 7-4 0-5 42-2 25 4-4 2-6 0

(B) Between median (distal segment) and saphenous nerves' Age Mean Sensory conduction Sensory conduction Distance Amplitude of No. recor (yr) age veloc. first finger- time first finger- (cm) sens. responses dable sens. wrist or knee-ing. wrist or knee-ing. at wrist or at potentials ligament (mlsec) ligament (msec) ing. ligament (A V) (no.ofnerves) on October 2, 2021 by Median nerve 21-52 35 49-2 7-4 2-6 + 05 12-4 i 03 23-1 10-9 0 Saphenous nerve 21-52 35 587 3-1 6,8 +05 39-2 + 2-1 56 3-3 0

All values are mean ± S.D. 'Eight normal subjects. 'Seven normal subjects. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.32.6.530 on 1 December 1969. Downloaded from 534 Cumhur Ertekin the amplitude of the saphenous sensory response TABLE III was higher than the response in the posterior tibial NUMBER OF PATIENTS IN DIFFERENT DIAGNOSTIC GROUPS nerve despite the longer distances involved in the saphenous nerve. Certainly, the mean amplitude was Poly- No clinical Total neuropathy poly- found significantly higher in the saphenous nerve neuropathy than in the posterior tibial nerve (P< 0005). The Diabetes mellitus 10 12 22 sensory conduction velocity was also faster in the Chronic alcoholism 6 11 17 saphenous nerve than in the posterior tibial nerve Patients with diseases associated with or without polyneuropathy: (P< 0-001). Renal insufficiency 4 3 7 To compare the sensory conduction in the Polyangeitis nodosa 2 1 3 saphenous and in the median nerves, seven un- Porphyria 1 - 1 Guillain Barr6 1 - 1 selected normal subjects were examined. The distal Carcinoma of lung 1 - 1 segment of the median nerve did not show a signi- B12 deficiency - 1 1 Lead intoxication - I 1 ficantly slower sensory conduction than the proximal Patients with unknown aetiology 11 - 11 segment of the saphenous nerve (P > 0 05). However, the of sensory responses amplitudes the in the median 36 29 65 nerve were significantly higher than those in the saphenous nerve (P<0001). The mean motor conduction time to the vastus the remaining 21 showed measurable potentials. medialis muscle was 5 05 msec in the younger group The mean sensory conduction velocity was 49-7 (mean distance 28-8 cm) and 5-08 msec in the older m/sec, which is slower than in the normal subjects group (mean distance 27-5 cm). Variations were 4-0 (P<0001). Seven cases, however, had velocities guest. Protected by copyright. to 6-0 msec in young (distances 21-7 to 34-5 cm) faster than 52 m/sec-that is, within the normal and 4 0 to 6-2 msec in old normal subjects (distances range. The amplitude at the inguinal ligament was 20-6 to 32-6 cm). There was a slightly longer motor also significantly reduced with a mean of 1-3 ,uV conduction time in the old than in the young group (P< 0o001). (P< 0 001), while the distances did not differ The femoral motor conduction time was prolonged significantly. The distance between the inguinal in a number of the cases with the mean values of ligament and the recording sites in the muscle 7-2 msec ranging 4-0 to 19-0 msec (P<0001). affected the motor conduction time, distances greater Prolonged femoral motor conduction time was than 32 cm gave latencies greater than 5 5 msec, usually associated with slow sensory conduction while those less than 25 cm had latencies less than velocity or absence of sensory response, whereas a 5 0 msec. normal motor conduction time was found both in cases showing slow and normal sensory conduction. PATIENTS In Table III, the patients examined are Thirty-four of these 36 cases were also investigated listed with their clinical diagnoses. Table IV displays for sensory conduction in the distal part of the the results for mean sensory conduction velocities median nerve. In four cases no median sensory in the saphenous and the distal part of the median action potentials were recorded. The remainder had nerve, as well as the results of the motor conduction a mean of medial nerve sensory conduction velocity measurement in the femoral nerve. of 38-3 m/sec and a mean amplitude of 6-1 ,tV. Thirty-six cases with clinically manifest poly- Both values are significantly smaller than in normal neuropathy were investigated. In 15, there were no subjects (P<0 001) compared with the normal recordable potentials at the inguinal ligament, while values at 40 to 61 years from the study of Buchthal http://jnnp.bmj.com/ ILE IV MEAN VALUES DETERMINED FROM THE INVESTIGATION OF 65 PATIENTS

Saphenous nerve Femoral nerve Median nerve Mean Nerves Sens. Ampl. of No Nerves Mot. Distance Ampl. of Nerves Sens. Ampl. of No age exam. cond. vel. sens. resp. exam. cond. (cm) muscle exam. cond. vel. sens. resp. (no.) knee- resp. (no. of (no.) time to resp. (no.) 1. fing.- resp. (no. of ing. lig. (MV) nerves) vast. med. (mV) wrist (V) nerves) on October 2, 2021 by (mlsec) (msec) (mlsec) Clinical poly- 51 36 49 7±4 7 1-29±0-5 15 36 7-20±2-6 28-8±2-4 11-3±9-3 34 38-347-5 6-1±4-4 4 neuropathy Subclinical 36 29 54-5±3 6 1-9±0-9 0 29 5-47±0-8 28-5±3-7 18-0±t1C5 24 43-8±7-2 13-7±6-8 0 cases All values are mean ± S.D. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.32.6.530 on 1 December 1969. Downloaded from

Saphenous nerve conduction in man 535 and Rosenfalck (1966). However, in 12 of 34 cases, Willison (1961), describing nerve action potentials sensory conduction velocity was still within the from the lateral popliteal nerve in man, the authors normal limits, although an amplitude reduction was were not able to record sensory action potentials at observed in some of them. the ankle with stimulation of the digital nerves in Twenty-nine patients who all suffered from the . Instead, they investigated mixed nerve diseases which sometimes affect the peripheral action potentials recorded at the head of the fibula nerves, but in whom careful clinical examination did after stimulation of the nerve at the ankle. Similarly, not reveal signs of polyneuropathy, were investigated Mayer (1963), and Mayer and Mawdsley (1965), (sub-clinical group). All patients showed sensory were rarely able to record sensory action potentials potentials at the inguinal ligament. The mean at the ankle on stimulation at the foot for either conduction velocity was 54 5 m/sec. These values posterior tibial or common peroneal nerves. are significantly lower than the values obtained for Similar difficulties were reported during attempts corresponding normal age groups (P< 0001). The to record sensory action potentials regularly from mean amplitude for the group was 19 ,uV, a signi- the in normal human subjects (Deaton ficant reduction from normal values (P< 0001). The and Downie, 1967). mean motor conduction latency in the femoral By introducing new techniques to record very nerve was 5 47 msec, which is moderately prolonged small action potentials using either an averaging (P< 0-01). computer (Dawson, 1954), a barrier grid storage Twenty-four of these 29 cases were investigated tube (Jensen, 1965; Gilliatt, Melville, Velate, and for sensory conduction in the distal part of the Willison, 1965), or a step-up input transformer median nerve. A mean sensory conduction of (Buchthal and Rosenfalck, 1965, 1966), it is possible

43-8 in/sec was found, which is slower (P<0005) to mneasure the sensory conduction velocities in the guest. Protected by copyright. than in young normals. The value is, however, not leg nerves. Thus, conduction velocities have been significantly different from middle age normals obtained in the posterior tibial nerve (Mavor and (P> 005) (Buchthal and Rosenfalck, 1966). The Atcheson, 1966) and in the and its mean amplitude was 13-7 ,uV, which was signi- distal branches (Buchthal and Rosenfalck, 1966). ficantly decreased (P

A comparison of the sensory conduction velocities 63 years. The responses could be recorded in some http://jnnp.bmj.com/ in the saphenous and in the median nerves showed patients over 65 as well (two of three cases). that the sensory conduction velocity was reduced in 2 A comparison in eight normal subjects with the saphenous nerve in 85 % and in the median the mean age of 45 years ranging from 33 to 59 nerve in 65 % of the cases with manifest poly- showed that the saphenous nerve is an alternative neuropathy. In the subclinical group, 31 % of the for measurement of sensory function in the lower patients had a sensory conduction velocity in the lim s. In all cases, the sensory action potentials in saphenous nerve below the lower normal limit, the saphenous nerve could be evoked at greater while 29 % had a slow sensory conduction velocity amE litude than in the posterior tibial nerve. on October 2, 2021 by in the median nerve (see Figs. 6, 7, 8 for individual 3. The method does not require an averaging recordings of the patients). cor iputer. S andardization of the position of the needle DISCUSSION electrode at the inguinal ligament was made by determining the threshold for stimulation of the NORMALS In the study of Gilliatt, Goodman, and motor fibres of the femoral nerve. This method has J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.32.6.530 on 1 December 1969. Downloaded from

536 Cumhur Ertekin

SAPHENOUS NERVE so So SO so SD So PNM: 17.t3.S 7.0 pM: 46.6±2.5 5.0 PN: S0.9t1.7 3.6 PNM: .1S1.3 3.6 S16 t 0.4 2.3 ST.1 ± O.S 2.3 M/see Sbc: 51911.0 3.6 SUbC:Uigl.1 If SUbe:U.6±t1.3 33 66r-

62 I- 0 0s 0 so I-. 0 0 0 00 0 0 A pO 0 I- a 14 0 0 0 S 50o - 000 .0 0 8 0 0 I 46 - 0 guest. Protected by copyright. 0 0 42 F 0

oI 0 0 0 __ __ PENORAL NERVE SD SD SO so so SD PN:69S03 12 PN: 16*13 19 PN: 6.2a1A 4A PN: U 60.3 1.0 S.0aLO OS0 .0 t1 IS Subs: L72 0.T Subs: 5t* 0.2 09 Subc: 4.71 1 0.4 msee j 0 14.0

9.0 1- 0 6.0 1 0 0S 7.0 _ 0 0 060 * * 0300 http://jnnp.bmj.com/ 6.0 00000 F 0@@ 0 a 0 a CMOO0 0 0 8 0 0 4.0 @0 0

GROUP DIAETIES CHRONIC GROUP WITH WITH NORMAL S NORMALS on October 2, 2021 by MELLITUS ALCOHOLISM OMR ETILOrtY UNKNOWN (17-3gymrs) (41-63 ystr) E TIOLOGY FIG. 5. Individual values for sensory conduction velocity in the saphenous nerve and for motor conduction time in the femoral nerve plotted according to aetiological groups. Filled circles denote patients with clinical polyneuropathy (PN) and open circles patients without clinical polyneuropathy (Subc). Filled triangles are normal subjects. The values at the top ofeach column are mean ± S.E.M. and S.D. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.32.6.530 on 1 December 1969. Downloaded from

Saphenous nerve conduction in man 537 the disadvantage that the position of the saphenous investigated by Heinbecker, Bishop, and O'Leary nerve in relation to the femoral nerve might vary (1933). They exposed the nerve in a diabetic patient between subjects and thus the distance between the with gangrene of the foot, and stimulated it directly recording electrode and the saphenous nerve will without an anaesthetic, as well as after excision of the vary in accordance. This disadvantage could be nerve about half an hour later. In this way they overcome by placing the recording electrode so that found a fast component conducting at 100 m/sec, a a minimal stimulus through it will give a subjective slower one conducting at 25 m/sec, and, with sensory response, but this method was not used as stimulus strength sufficient to stimulate all fibres of it requires the co-operation of the patient. the nerves, an additional component conducting at Saphenous nerve conduction in man was first 1 5 m/sec. Such exposure techniques are clearly not applicable to routine clinical examinations. SaphenoLs Femoral Median rve nerve nerve FIG. 6. Patients with diabetes mellitus. Sensory action 44 potentials from the saphenous and median nerves recorded 60 27cm at the inguinal ligament and at the wrist after stimulation at the knee and first finger respectively (first and third A d\J?.~iuh J columns) and the motor responses recordedfrom the vastus 1] medialis muscle with stimulation at the inguinal ligament supramaximally (second column). All sensory responses 26 cm 45 are 20 superimposed potentials. Figures on each trace for 50 sensory responses give the conduction velocity in m/sec and figures on each trace for motor responses show the B ] sites in the 04%0- ] distances between stimulation and recording guest. Protected by copyright. muscle. A. E. W., 23-year-oldfemale (467/67, 1 September 1967). Duration of diabetes is six weeks. No clinical 48 32.5 cm 34 neuropathy. Normal values for three nerves. B. M.H., 27-year-old female (3/68, 2 Janaury 1968). Duration of C I ] diabetes is nine years, no clinical neuropathy. Slow sensory m y ] conduction in the saphenous nerve, normal values in the femoral and median nerves, while the amplitude of the sensory responses at the wrist is reduced. C. F. W., 29-year- 27 cm 27 old male (489/67, 11 September 1967). Duration of I diabetes is 10 mild Slow * J years. Clinically neuropathy. D I] ^_/W 6apV sensory conductions in both nerves with prolongedfemoral nerve motor conduction time (7.0 msec). D. W.S., 59-year- r-- old male (408/67, 7 August 1967). Duration of diabetes is about one year. Clinically sensory-motor symmetrical 5msec polyneuropathy with subacute onset. No certain sensory response in the saphenous nerve and slow sensory con- Saphenous Femoral duction in the median nerve. Femoral nerve conduction nerve nerve nerve time is also prolonged (8-0 msec). Vertical scales indicate 53 35cm 55 6 m V, horizontal scale 5 msec. FIG. 7. Patients with chronic alcoholism. Same arrange- A I ment of columns as in Fig. 6. A. L.H., 27-year-old male (710/67, 23 November 1967). Heavy drinking every dayfor http://jnnp.bmj.com/ four years. One month earlier mental symptoms appeared. No clinical neuropathy. Values for median and femoral 47 nerves are within the normal limits, but sensory conduction 52 31 cm in the saphenous nerve is below the normal range of the younger normal subjects. B. J.K., 29-year-old male (692/67, I *,~\> 20 November 1967). Heavy drinker everyday for seven B ] nfr) years. Hepatic cirrhosis is present. Clinically no neuro- pathy. The value for sensory conduction in the saphenous

nerve is below the normal rangefor younger normal subjects on October 2, 2021 by 49.5 32 cm 33 while the values for the median and the femoral nerves are normal. C. G.N., 62-year-old male (796/67, 27 December * A ]6uV 1967). Many years of alcoholism. Clinically mild poly- C 1 neuropathy. The values for sensory conduction in both nerves are slow, but femoral conduction time is 6-5 msec. Vertical scales indicate 6 m V. Horizontal scales indicate 5msec 5msec. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.32.6.530 on 1 December 1969. Downloaded from

538 Cumhur Ertekin

Femoral Median both young and older subjects reported in the liter- nerve nerve nerve ature (Buchthal and Rosenfalck, 1966; Kemble and 53 22cm 50 Peiris, 1967). With increasing age, the sensory con- duction velocity was found to be slower in the A i -;3] saphenous nerve, a finding that agrees with other (VP ~ studies on other sensory nerves (Downie and Newell, 1961; Mayer, 1963; Buchthal and Rosenfalck, 1966). 46 31cm 44 The distal segment of the saphenous nerve had a lower conduction velocity than that of the proximal B d segment determined in 10 young normal subjects. This is also similar to observations in other mixed and sensory nerve conduction studies (Magladery 36 and McDougal, 1950; Mayer, 1963; Buchthal and Rosenfalck, 1966; Kemble and Peiris, 1967). cm~] t _/4] The findings for the femoral conduction time in normal subjects is slightly shorter than in the studies U- of Gassel (1963) and of Chopra and Hurwitz (1968). 49 29 cm 36 The slightly different values obtained in the present study despite the comparable distances were prob- D ] 4\_-.%J 1 6pV ably due to use of the vastus medialis muscle in contrast to the other two studies mentioned above,

._J L ------J where the evoked muscle action potentials were re- guest. Protected by copyright. s msec corded from the rectus femoris muscle. FIG. 8. Patients with renal insufficiency aand polyangeitis nodosa. Same arrangement of columns cis in Fig. 6. A. PATIENTS In the patients with and without signs of A.G.S., 51-year-old female (20/68, 20 Alovember 1968). clinical polyneuropathy the sensory conduction in Bilateral renal cyst for five years with exacerbation of the saphenous nerve differed from the normal chronic pyelonephritis and renal insuffici!ency. Clinically values. These changes in sensory conduction occurred no neuropathy. The values for all three n both and absence of the normal limits, but the sensory cotIduction in the in presence corresponding saphenous nerve is close to the lower range ofnormal older disturbances in the median sensory and in the femoral subjects. B. L.N., 31-year old male (695/6'7, 20 November motor nerves. 1967). Three years chronic renal insufficieincy with nephro- The fact that 85% of the cases with clinical lithiasis. Clinically mild polyneuropathy. Irhe sensory con- polyneuropathy have either absence of sensory duction in the saphenous nerve is slow, butt still within the response at the inguinal ligament or less than 52 normal limits in the median nerve, althoui gh the amplitude m/sec conduction velocities in the saphenous nerve of the sensory response at the wrist reduce d. Femoral con- shows that the nerve is frequently involved in poly- duction time is also prolonged (7 0 msec). CX M.J., 22-year- neuropathy; while the femoral conduction time of old male (706/67, 22 November 1967). S year 7 0 msec or more was found in about 39 % of all gressive renal insufficiency with renal cysiFven under dialysis. Clinically severe sensory aand motor poly- cases with neuropathy. Therefore the saphenous neuropathy. No recordable sensory potentials in the nerve conduction seems to be affected more than saphenous nerve and slow sensory conduction in the median twice as often as that of the femoral nerve. These

nerve. Femoral conduction time is extremely prolonged findings are in agreement with the pathological http://jnnp.bmj.com/ (14*0 msec): note the slow sweep speed. D. IB.T., 45-year-old findings that changes in sensory fibres occur before male (31/68, 17 January 1968). Polyeangeitis nodosa those in motor fibres in different kinds of neuro- diagnosed 2 5 years earlier. Clinically mildpolyneuropathy. pathies (Gilliatt and Thomas, 1960; Gilliatt, 1961; Slow sensory conduction velocities in theosphenos and Gilliatt et al., 1961; Gilliatt and Willison, 1962; the median nerve, while normal femoral conuction time. and such Vertical scales indicate 6 m V, horizontal scales 5 msec. Downie, 1964; others). Although para- meters as amplitude and duration of the sensory The mean sensory conduction ve:locity of 59-6 potentials recorded at the wrist in the median nerve

m/sec in the proximal segment of 1the saphenous deviate significantly from normal values and limits, on October 2, 2021 by nerve in young normal subjects is siniilar to that in sensory conduction velocity in the distal part of the the proximal segment of the sciatic:nerve between median nerve below the borderline of the normal or fossa poplitea-buttock with the mean of 62A4 m/sec no recordable sensory potentials at wrist was found as determined by Buchthal and Rosenfalck (1966); in 65 % of the cases with polyneuropathy in contrast and considerably slower than that ini the proximal to 85 % of the cases in the saphenous nerve. This segments of the median nerve (elbovw to axilla) in greater incidence of deviations from normal limits J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.32.6.530 on 1 December 1969. Downloaded from

Saphenous nerve conduction in mran 539 in sensory conduction velocity in the saphenousnerve including diabetes, it would be interesting from than in the median nerve would parallel the clinical conduction studies to learn whether the cases with findings that in many cases leg nerves were involved this syndrome display mainly a motor neuropathy more than arm nerves. or involve both sensory and motor fibres. In 29 subclinical cases associated with different diseases, there was no sign that could be interpreted SUMMARY as evidence of polyneuropathy, yet the saphenous nerve conduction velocity was significantly lowered 1. A method is described of recording sensory in 31 % of the cases, being below the lower limit of potentials in the saphenous nerve in man. the normal range. While two cases had a femoral 2. The suitability of this nerve for clinical sensory motor nerve latency of 7 0 msec, the mean of the conduction velocity measurement in the leg is dis- femoral conduction time was significantly prolonged. cussed. The sensory conduction in the distal part of the 3. It is shown that the saphenous nerve may median nerve was almost equally involved in 29 % display early changes in sensory conduction in of cases but was statistically closer to normal values patients without clinical polyneuropathies as well as than that of the saphenous nerve. Slow sensory con- frequent involvement of the nerve in patients with duction found in such subclinical cases in different clinical polyneuropathies. nerves has been reported previously (Downie and 4. More severe affection of sensory conduction in Newell, 1961; Mayer, 1963; Preswick and Jeremy, the saphenous nerve in combination with or without 1964; Mawdsley and Mayer, 1965; Jebsen,Tenckhoff, prolonged femoral nerve conduction may occur in and Honet, 1967; and others). diabetes more frequently than in other diseases

The method for measurement of the sensory con- associated with neuropathy. guest. Protected by copyright. duction in the saphenous nerve appears to be of value to reveal subclinical changes in sensory nerve I am grateful to Dr. D. H. Ingvar for providing excellent fibres in the leg. Furthermore, as is stated by Chopra working facilities and for guidance and encouragement and through all phases of the present study. I also wish to and Hurwitz (1968), both femoral saphenous express my sincere gratitude to Dr. D. Elmqvist for help, nerves are less liable to compression and trauma advice, and constructive criticism. anatomically than the median, ulnar, and lateral popliteal nerves. Therefore in such situations, REFERENCES measurement of the saphenous nerve conduction provides a more useful clinical aid than studies of the Buchthal, F., and Rosenfalck, A. (1965). Action potentials from sensory nerve in man; physiology and clinical application. other nerves mentioned, especially in subclinical Acta neurol. scand., 41, Suppl. 13, 263-266. polyneuropathies. -,1 (1966). Evoked action potentials and conduction velocity in human sensory nerves. Brain Res., 3, 1-122. The material presented in this study is not sufficient Calverley, J. R., and Mulder, D. W. (1960). Femoral neuropathy. for valid comparisons between disease of different Neurology (Minneap.), 10, 963-967. aetiology, such as diabetes, chronic alcoholism, Chopra, J. S., and Hurwitz, L. J. (1968). Femoral nerve conduction in diabetes and chronic occlusive vascular disease. J. Neurol. uraemia, but in diabetes the saphenous and femoral Neurosurg. Psychiat., 31, 28-33. nerves seemed to be more frequently involved. Dawson, G. D. (1954). A summation technique for the detection of small evoked potentials. Electroenceph. clin. Neurophysiol., These findings are in agreement with the clinical 6,65-84. observations in diabetes mellitus ofHirson, Feinmann, Deaton, P., and Downie, A. W. (1967). Quoted by Downie, A. W., and Scott, T. R. (1967). An improved technique for radial nerve and Wade (1953); Goodman (1954) and Calverley conduction studies. J. Neurol. Neurosurg. Psychiat., 30, and Mulder (1960), and of the electrophysiological 332-336. http://jnnp.bmj.com/ studies of Chopra and Hurwitz (1968); Garland and Downie, A. W. (1964). Studies in nerve conduction. pp. 511-535. In Disorders of Voluntary Muscle, 1st edn., edited by J. N. Walton, Taverner (1953) ascribed diabetic amyotrophy to a J. and A. Churchill: London. condition affecting the spinal cord, but Gilliatt and and Newell, D. J. (1961). Sensory nerve conduction in patients with diabetes mellitus and controls. Neurology (Minneap.), Willison (1962), in an electrophysiologcal inves- 11, 876-882. tigation, concluded that it was a motor neuropathy. Garland, H., and Taverner, D. (1953). Diabetic myelopathy. Brit. came con- med. J., 1, 1405-1408. Chopra and Hurwitz (1968) also to this Gassel, M. M. (1963). A study of femoral nerve conduction time. clusion by their conduction studies in the femoral Arch. Neurol. (Chic.), 9, 607-614. nerve in diabetics and chronic occlusive vascular Gilliatt, R. W. (1961). Nerve conduction: motor and sensory. pp. on October 2, 2021 by 385-411. In Electrodiagnosis and Electromyography, 2nd edn., disease. Unfortunately, there was no case which had edited by S. Licht. Licht: New Haven, Connecticut. the features resembling diabetic amyotrophy in the Goodman, H. V., and Willison, R. G. (1961). The recording of lateral popliteal nerve action potentials in man. J. Neurol. present material, but, since the sensory conduction Neurosurg. Psychiat., 24, 305-318. in the saphenous nerve as a terminal sensory branch Melville, I. D., Velate, A. S., and Willison, R. G. (1965). A study nerve was more as of normal nerve action potentials using an averaging technique of the femoral affected than twice (barrier grid storage tube) J. Neurol. Neurosurg. Psychiat., 28, often as that of the femoral nerve in neuropathies 191-200. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.32.6.530 on 1 December 1969. Downloaded from

540 Cumlur Ertekin

and Sears, T. A. (1958). Sensory nerve action potentials in Jensen, A. S. (1955). The radechon, a barrier grid storage tube patients with peripheral nerve lesions. J. Neurol. Neurosurg. RCA Review, 16, 197-215. Psychiat., 21, 109-118. Kemble, F., and Peiris, 0. A. (1967). General observations on sensory -, and Thomas, P. K. (1960). Changes in nerve conduction with conduction in the normal adult median nerve. Elec;romyo- ulnar lesions at the elbow. J. Neurol. Neurosurg. Psychiat., 23, graphy, 7,127-140. 312-320. Magladery, J. W., and McDougal, D. B. (1950). Electrophysiological -, and Willison, R. G. (1962). Peripheral nerve conduction in studies of nerve and reflex activity in normal man. Bull. Johns. diabetic neuropathy. J. Neurol. Neurosurg. Psychiat., 25, 11-18. Hopk. Hosp., 86, 265-290. Goodman, J. I. (1954). Femoral neuropathy in relation to diabetes Mavor, H., and Atcheson, J. B. (1966). Posterior tibial nerve con- mellitus; report of 17 cases. Diabetes, 3, 266-273. duction. Archesol. B.).14 661ne69oi Heinbecker, P., Bishop, G. H., and O'Leary, J. (1933). Pain and touch duction. Arch. Neural. (Chic.)., 14, 661-669. fibers in peripheral nerves. Arch. NeLrol. Psychiat., 29 771-789. Mawdsley, C., and Mayer, R. F. (1965). Nerve conduction in alcoholic Hirson, C., Feinmann, E. L., and Wade, H. J. (1953). Diabetic polyneuropathy. Brain, 88, 335-356. neuropathy. Brit. med.J., 1, 1408-1413. Mayer, R. F. (1963). Nerve conduction studies in man. Neurology Jebsen, R. H., Tenckhoff, H., and Honet, J. C. (1967). Natural history (Minneap.), 13, 1021-1030. of uremic polyneuropathy and effccts of dialysis. New Engl. Preswick, G., and Jeremy, D. (1964). Subclinical polyneuropathy in J. Med., 277, 327-333 renal insufficiency. Lancet, 2, 731-732. guest. Protected by copyright. http://jnnp.bmj.com/ on October 2, 2021 by